Utilities are often supplied from underground lines. Two techniques are generally used to install such lines. In one technique, the utility line pathway is excavated; the line is installed; and the excavated material is replaced. While this method is suitable for new developments, implementation of this technique is not always practical in previously developed areas. As a result, industry development efforts have been focused on excavating tools capable of installing utilities underground without surface disruption.
Several guided and unguided boring tools are currently on the market. Guided tools require substantially continuous location and orientation monitoring to provide the necessary steering information. A prerequisite of such monitoring is, of course, locating the tool that is to be monitored. Only once the position of the tool is located can a proper depth measurement be obtained, for example, from a measuring position directly above the head of the boring tool which houses a transmitter. Unguided tools would also benefit from periodic locating or substantially continuous monitoring, for example, in prevention of significant deviation from planned tool pathways and close tool approaches to utilities or other below surface obstructions.
Locating or monitoring systems currently used in combination with boring apparatus are either cable locating systems or are based on cable locating technology. Although the more advanced systems perform adequately, limitations on cable locating technology also limit measurement accuracy.
Most cable locators involve receiver detection of an oscillating magnetic field derived from electrical current directly fed or induced onto the cable. The magnetic field lines emanating from a cable are essentially cylindrical in shape, forming concentric circles around the cable. As the current flows along the cable, losses occur as a result of displacement and induced currents into the soil. Consequently, the exact signal strength of the magnetic field emanating from the cable at any point is unknown. Although local signal peaks or nulls (depending on receiver antennae and electronic configuration) are useful to determine the surface position directly above the cable, signal strength (i.e., magnetic field strength) alone is not directly indicative of cable depth. In certain specific circumstances (i.e., when the rate of loss along the cable length is not great), a signal strength ratio can be used to compute depth. If the cable run is straight for a long distance (compared to the depth), the magnetic field strength (B) will be inversely proportional to the distance (d) from the cable to the receiver (i.e., B .alpha.1/d or B=k/d, where k is a proportionality constant). By taking two signal strength readings at different locations directly above the cable, the proportionality constant can be eliminated and the depth determined.
A simple device for determining the depth of a relatively straight cable is manufactured by Dynatel, a subsidiary of the Minnesota Mining and Manufacturing Company. The Dynatel device includes a single antenna, a gain control knob and a gain doubling switch. The operator determines cable depth by (1) placing the device on the ground above the cable; and (2) adjusting the output displayed on a meter with the gain control knob until the meter needle lines up with a line on the meter scale; (3) doubling the gain with the switch therefor; and (4) vertically elevating the device until the output returns to the original value (i.e., the needle realigns with the meter line referred to in step (2)). Since the magnetic field strength is inversely proportional to the distance, the height of the unit above the ground at step (4) is equal to the depth of the cable. This procedure is accurate, but time consuming. It also becomes impractical for more deeply buried cables, requiring the operator to raise the device above his head.
Other currently used cable locating devices employ two antennae and logic circuitry to determine depth. The antennae are separated by a fixed distance. With this known separation distance and magnetic field strength readings at the antennae, cable depth can be computed. The difficulty with these devices is that there are practical limits regarding antennae separation. If the cable depth is much larger than the antennae separation, which is generally approximately 12 to 18 inches, signal strength measurement accuracy becomes more critical. Measurement accuracy is affected by differential drifting of the electronics associated with the antennae as well as differential responses of the antennae themselves.
Various approaches have been taken to enhance magnetic field strength measurement precision. The accuracy of these approaches increases as the number of components common to the two measurement circuits increases. Current systems accomplish this by taking a magnetic field reading at one antenna; switching the electronics connection from one antenna to the other; and measuring the magnetic field strength at the second antenna. Although this switching methodology eliminates many sources of error, one major error source remains--the antennae. To increase sensitivity, ferrite rods are sometimes employed to enhance the effective capture area of the antennae. As a result of the antennae separation, both antennae may not experience the same thermal environment. The characteristics of ferrite vary measurably with temperature and are not consistent between rods. Alternatively, large diameter air-core coils are employed. Such coils eliminate the inconsistency of the ferrite rods, but still exhibit thermal drift problems. Air-core coils also are generally larger in diameter.
All of these spatially separated two-antenna devices must be periodically calibrated. Any aging or drifting of an antennae will cause rapid loss in cable depth measurement accuracy, particularly at depths that are large compared to antennae separation. In cable locators, this is generally not a serious problem, since most cables are buried at depths of less than 2 or 3 times the separation.
A device conforming to the above-described arrangement is available from Radiodetection Ltd. (Bristol, England), the RD300. The device includes two antennae with horizontal coil axes disposed a fixed vertical distance from each other. In operation, the device is placed on the ground, such that a first receiving antennae sensor is near ground level (e.g., within about 1-2 inches) and a second receiving antennae is located about 16 inches thereabove. The ground therefore serves as a reference surface for depth measurement. One disadvantage of this particular prior art device and other devices that operate similarly thereto manifests itself when the reference surface exhibits an obstruction such as a curb, a rock, landscaping or the like, at a desired measurement location. Under these circumstances, an operator must compensate for the obstruction to obtain the depth below the reference surface. Another disadvantage of this equipment is that the depth measurement process is time consuming even after the device is properly located above the transmitter (i.e., a needle must be aligned with a meter line through a knob-actuated adjustment process). Radiodetection Ltd. applies this technology to cable, sewer and pipe location as well as horizontal boring tool monitoring.
The principal means of locating a boring tool head for guidance purposes is to place a radio frequency transmitter in the tool head, and track the tool from the surface using a radio frequency receiver that detects the alternating magnetic field emanating from the transmitter. While this is similar to the cable-locating situation, the type of measurement necessary for accurate guided boring differs, and the requirements there for are more stringent. Transmitters or sondes generally emit a dipole magnetic field in the normal measurement range, which differs from the source or source-like magnetic field emanating from a utility cable. When a single horizontal antennae is used to measure the strength of a dipole magnetic field, that parameter varies as depicted in FIG. 1a.
A transmitter 10 is located directly below a maximum field strength point 12. Nulls 14 are present in the horizontal field directly ahead and behind maximum 12, causing local peaks 16 in field strength. If a locator/monitor operator were to commence operations at a location substantially ahead or behind the actual transmitter 10 location, he might locate one of local peaks 16 and believe the tool to be directly below. In order to be certain that field strength maximum 12 has been located when using single horizontal antennae devices, another peak must be found and evaluated to be lower in strength (i.e., to be a local peak 16). An operator failing to take this precautionary measure may conclude that transmitter 10 is located at a position that leads or trails its true location. Erroneous depth readings and subsequent misplacement of the bore typically result.
A single vertical antennae fares no better. A vertical antennae will produce a null directly above the transmitter. This null exists along a line extending on both sides of the transmitter, however, and therefore cannot be used to locate a point, such as the transmitter location. Data from a combination of two antennae may be manipulated to provide a more accurate indication of transmitter location. An orthogonal set of antennae can produce the monotonic signal strength variation shown in FIG. 1b.
When guiding a boring tool, the operator constantly requires accurate depth measurements, and time consuming procedures, such as the single antenna cable locator utilizing gain doubling, are therefore not practical. For tool control purposes, the operator must be able to determine the depth gradient to ascertain the direction (i.e., up or down) in which to steer. Gradient determinations require greater precision than depth measurement. Also, boring depth may be a factor of 0 or more greater than practical antennae separation limits of spatially separated two antennae locators.
U.S. Pat. No. 4,806,869 issued to Chau et al. discusses a 5-sensor receiver apparatus capable of "locating the position of a boring device within the ground with respect to a particular reference location along an above ground path directly over the intended course" of the boring device. In this receiver, four sensors are arrayed at the four corners of a square within a horizontal plane (i.e., parallel to the surface), the midpoint of which is displaced vertically from the fifth sensor. Chau et al. indicate that such a receiver is an improvement over a 4-sensor device designed to locate/monitor electronically conductive cable, having sensors located at the end points of two intersecting lines of equal length within a plane that is perpendicular to the surface.
The 4-sensor cable-locating apparatus was not designed for continuous monitoring. Signals from the horizontally placed sensors are used to locate the transmitter, while signals from the two vertically aligned sensors are used to determine cable depth. Such a process is impractical for continuous monitoring.
In contrast, the 5-sensor apparatus utilizes signals from the two horizontally disposed sensors, located in the plane perpendicular to the desired path of the boring device and within which the boring device is actually positioned, and the vertically displaced sensor to determine boring device depth and displacement from its intended path.
The disadvantage of the 5-sensor device is its complexity. This device is also susceptible to locating local peaks 16 in the signal strength. Also, the operator of a 5-sensor device traverses the desired boring device path, rather than locating a position directly above the device.
Again, these 4- and 5-sensor prior art receivers incorporate sensors that are in fixed spatial positions with respect to each other. In contrast, U.S. Pat. No. 4,646,277 issued to Bridges et al. includes a sensing assembly formed of three orthogonal pick up coils. The sensing assembly of the Bridge et al. patent serves as a homing beacon for a boring apparatus, rather than a means to establish the position of the tool head.
U.S. Pat. No. 3,906,504 issued to Guster et al. describes a method of locating and plotting tunnels using a portable receiver to monitor a transmitter moving through the tunnels. Guster et al. employ an antennae having a vertical axis in the transmitter. While this antennae configuration eliminates nulls, such an arrangement is not practical in a boring application, because the head of the boring apparatus rotates. Signal strength emanating from a vertically oriented antennae would therefore vary during boring.
Also, Guster et al. employ very complex mathematics in determining the distance between the transmitter and the receiver. The need for a calibration system involving complicated electronics for use with the Guster et al. system is discussed, without further explanation, at Column 2 of the patent. The Guster et al. estimate regarding the complexity of calibration electronics appears to be accurate in view of the nature of the depth determination employed in the patent.
In addition, Guster et al. employ a pulsed transmitted signal, so as to avoid interference with verbal communication between the receiver operator and the transmitter operator. Pulsed transmitted signals complicate the locating/monitoring process carried out by the receiver.
Steering a boring device also requires information concerning pitch (i.e., angle above or below the X-axis in an XY plane, where the X-axis corresponds to the longitudinal axis of the boring device and the Y-axis is parallel to the gravity vector). Several pitch sensors are known and commercially available. Most of these pitch sensors will not produce a pitch angle independent of the roll orientation (about the X-axis). Those that can produce a roll-insensitive signal are generally expensive to produce and easily damaged by shock loads. Less expensive pitch-sensing devices are generally not sufficiently sensitive or well damped. Because equipment loss is common, most users are reluctant to invest a large amount of money in components that are deployed underground. Consequently, development of low cost pitch sensors capable of surviving the loads and environment associated with boring through soil, rock and debris has been pursued.
U.S. Pat. No. 4,674,579 issued to Geller et al. describes two pitch-sensing devices. One apparatus features a transmitter that includes a mercury switch connected in such a manner that the transmitter is deactivated when the tip of the housing therefor is upwardly inclined. The inclination of the tip may be determined by an operator by measuring the angle of rotation at which the transmitter switches on and off. This type of pitch-sensing device is not highly accurate as a result of inaccuracy in measuring the roll angle of the tool head. This process is also time consuming, thereby reducing the practicality of implementing such a methodology.
The second pitch-sensing device shown in FIG. 8 of and described in the Geller et al. patent includes a first common electrode and two pad-electrode assemblies, including the second and third electrodes, housed within a glass envelope. The glass tube is partially filled with an electrolytic fluid, such that the resistance between the second and third electrodes and the first common electrode varies with the inclination (i.e., pitch) of the device. This pitch-sensing device can be costly to implement.
An additional difficulty with locating and monitoring boring apparatus having a transmitter housed in the boring tool head is that the structural loads and wear experienced by the tool head require that the head be fabricated from a high strength material such as steel or some other metal. Since metals conduct electricity, a transmitter contained within a metal tool head induces a current in the metal. This induced current, in turn, induces a magnetic field that cancels the transmitted field to some extent and, in some circumstances, entirely.
In order to allow the signal emitted by the transmitter to radiate to the surface, one or more windows or openings have been fabricated or machined into the conductive boring tool head. Employing this solution structurally weakens the tool head and may allow debris or ground water to enter the tool head and impinge upon the transmitter, thereby destroying the antennae and/or the related electronics. To avoid such debris and water damage and in an effort to bolster the strength of the windowed tool head, these openings have been filled with composite, ceramic or plastic materials, thereby sealing the transmitter and antennae. These filler materials are not as durable as metal, however, and generally fail long before a metal structure would fail. Typically, filler material failure results in costly electronics destruction. Since the tool structure is weakened by the window, premature tool head failures resulting in the loss of both the tool head and the electronics may also occur, however.
Another difficulty with the use of the window concept is that the radiated field strength becomes a function of tool head orientation. Specifically, in a single window configuration, the field is strongest when emanating from the window and measurably weaker 180.degree. therefrom. Although this result can be useful in determining the tool head roll orientation, it makes it impossible to determine tool depth accurately while drilling, because the tool head is rotating during drilling. To overcome this restriction, multiple small window or slot tool head designs have also been used with mixed success.
In another attempt to overcome this radiated signal problem, the entire tool head structure has been formed with non-conductive materials such as composites and ceramics. Unfortunately, none of these substitute materials exhibits all of the desirable characteristics of steel or other durable conductive metals. Strong ceramics do not handle impact loads as well, while composites do not take abrasive wear as well. These substitute materials are also much more costly than metals.